It was a beautiful day, light snow/freezing rain falling ( here 2 hours north of Toronto) on Friday. As night fell the wind whipped up. I woke up in the morning and wondered why it felt cool. A quick check determined that the power was out. No doubt the wind and freezing rain had brought down a few power lines. Of course the first thing I did was call (thank goodness for old phones that draw power over the telephone line) Hydro One our local power company. Being surrounded by seasonal cottages I wanted to make sure the utility knew we were out of power.
The call was answered by a virtual agent, who stated they had a 'reported power outage in my area'- thank you calling line ID. Of course I don't know how fine tuned this power outage location was so I still wanted to make a report. Guided by the helpful virtual agent I identified myself and completed an automated ticket and the virtual agent terminated the call.
After a few hour of stoking the wood stove and fighting to maintain 50 degrees inside I thought I would call Hydro One again and try to get an update. the first five calls resulted in a "due to unexpectedly high call volume we cannot answer your call please call back', but on call six I succeeded and got in the door. I selected the speak to an agent option, and was placed in queue. Now I know one or two things about call and contact centers so I knew Hydro One was operating in a defined queue depth environment (hence the we are too busy call back message), so I knew that once I was in then there was an expected rough time line to answer. For example if they had 20 agents on who took 20 calls an hour and they wanted a maximum queue depth/delay of 30 minutes then they would set the queue depth to 200 calls. I know the power was out and I knew that I wasn't the only one affected so I expect a delay. I assumed that the queue depth management approach I felt they were employing would ensure a delay of 20 to 40 minutes. Boy was I wrong, after 90 minutes of silence; no queue messages, not even a thank you for holding, I abandoned. What I believe happened was that I was giving the call center management way too much credit. I had assumed they scaled the queue depth based upon the day of week or agents scheduled...but now I don't think so. What I think happened was that the center which may staff 250 agents Monday through Friday and only 40 or 50 on Saturday never adjusted the queue depth so my 30 to 40 minute estimate was likely off by 500%!
There are some lessons for those of us who manage call & contact centers including;
1- If you will report that you already know about a problem, provide enough detail to provide reassurance to customer that you really do know about the situation otherwise they will often call back for the reassurance,
2- Employ queue messages, we all want to know that we are still connected, but don't simply tell us we are important...this will get a little old and sound a bit ingenuous after 2- or 30 minutes.
3- You need manage your tools, simply setting up queue depth thresholds, or schedules based upon a normal day doesn't cut it. This isn't hard to do and ensuring that your have scaled based upon your available resources will ensure you provide better service.
The power did come back on after a 14 hour outage, so now we are toasty and warm. I wonder if anyone at Hydro One will examine their blockage reports on Monday and do a root cause analysis and come to the same conclusions I did...I doubt it.
No comments:
Post a Comment